In a series of papers dating from 1929, Hans Berger has described ‘a rhythmic oscillation of potential at a frequency of 10 cycles per second … detected in the human subject by electrodes applied to the head … present when the subject lies quietly with eyes closed and disappearing when attention is fully occupied’. Because Dr Berger finds that the waves are larger if needles reaching the periosteum lie over foramina in the skull rather than the intact cranium, he concludes that this remarkable effect derives from the cerebral cortex, and that it reflects a fundamental activity of the brain. But there are unresolved issues, and Professor (Lord) Adrian and Dr (Sir Bryan) Matthews intend to repeat Berger’s experiments: to show whether or not these potentials do indeed arise from the cortex; to reveal their nature and the structures from which they arise; to characterize the type of activity that they represent; and to reconcile the emerging results with their prior failure to record comparable activity from the exposed surface of the animal cortex. ‘To anticipate … we regard the effect as due to spontaneous rhythmic activity of … cortical cells in … the occipital lobe … [that] beat synchronously when … undisturbed … but visual activity or wide-spread non-visual activity in the brain breaks up the rhythm by exposing the cells to a mosaic of excitations which makes synchronous action impossible … localization is the only important point on which our results seem to differ [from Dr Berger].’ Their method is to detect potentials using lint-covered squares of copper gauze soaked in saline. These make good contact with the scalp and, contained by a tight head bandage, are not inferior to the implanted needle electrodes used by Berger. The subject remains quiet and physically relaxed during the recording. The traces are viewed optically on bromide paper using the Matthews oscillograph that registers events occurring at 1/6000 s, as well as being transduced into sound. Without much difficulty, Professor Adrian and Dr Matthews demonstrate the Berger rhythm, at an amplitude of 0.1 mV, present in the dark or with the eyes closed and disappearing on eye opening (Figs 1 and 2). But does it originate in the brain? A contribution from muscles of the face and neck seems unlikely since the regularity and rhythm of muscle contraction are altogether different from the 10/s Berger trace; nor are they influenced by active movements of either muscle group. The relation to eye opening and closure makes it more difficult to exclude artefact from the orbital muscles, and ‘for a time we suspected that the rhythm arose in this way’; but it would be difficult to capture orbital muscle activity through electrodes placed over the occiput and neither active nor passive movements of the eyeballs reproduce or influence the rhythm. Dr Berger was able to record activity over every part of the cortex, as can Adrian and Matthews, but—in considering its site of origin—they are much influenced by the requirement for an electrode placed just above the external occipital protuberance in order to maximize the amplitude of the recorded trace (Fig. 3). Between them, Dr Berger and the present authors have excluded, to their combined satisfaction, an origin for the rhythm from pulsation of vessels, activity of pilomotor muscles, tremor of the head, or retinal potentials. Pursuing Berger’s observation that the size of the potential directly reflects proximity of the recording electrode to the surface of the brain, Adrian and Matthews study five patients whose skull has previously been trephined. Two with temporal defects lose their Berger rhythm when the recording electrode is moved forwards from the occiput; they interpret this paradox in terms of the theoretical distribution of current recorded through an intact cranium and that seen in the context of a local defect—observations further reinforcing their conclusion that the potentials come from inside and not outside the skull (Fig. 4). Taken together, it seems certain that the rhythm must originate from the surface of the occipital cortex. However, its distribution shifts as the beat spreads progressively over the active area instead of appearing simultaneously at every point. It follows that at any one time, the rhythmic shift in the distribution of current that corresponds to the flow of maximum activity will give rise to phase differences in potentials recorded from different leads placed over the cortex, especially as these get further apart. That said, the rate and direction of flow seem to be matched across the two hemispheres, which do appear to activate synchronously. Records of the Berger rhythm made with pad electrodes on the head (vertex and occiput), with the subjects sitting with the eyes closed. (A) Subject E.D.A. Record made with Matthews’s oscillograph, period 1/6000 s. Maximum excursion 0.07 mV. Frequency 9.5/s. (B) Subject W.H. Record made with ink-writer oscillograph, period 1/40 s. Frequency of waves 9/s. (C) Subject K.Y. Record made as in B. Frequency 10.5/s. Note that all ink-writer records are at the same speed. Of the subjects examined by us, W.H. gives the slowest and K.Y. the most rapid rhythm. Records of the Berger rhythm made with pad electrodes on the head (vertex and occiput), with the subjects sitting with the eyes closed. (A) Subject E.D.A. Record made with Matthews’s oscillograph, period 1/6000 s. Maximum excursion 0.07 mV. Frequency 9.5/s. (B) Subject W.H. Record made with ink-writer oscillograph, period 1/40 s. Frequency of waves 9/s. (C) Subject K.Y. Record made as in B. Frequency 10.5/s. Note that all ink-writer records are at the same speed. Of the subjects examined by us, W.H. gives the slowest and K.Y. the most rapid rhythm. The development of the rhythm in the absence of visual activity. (A) E.D.A. The rhythm appears when the eyes are closed. (B) B.H.C.M. Same as for A. (C) E.D.A. The rhythm disappears when the eyes are opened. (D) W.H. After some minutes in the dark the rhythm is present with the eyes open. Closing them does not alter the rhythm. The development of the rhythm in the absence of visual activity. (A) E.D.A. The rhythm appears when the eyes are closed. (B) B.H.C.M. Same as for A. (C) E.D.A. The rhythm disappears when the eyes are opened. (D) W.H. After some minutes in the dark the rhythm is present with the eyes open. Closing them does not alter the rhythm. Comparison of waves recorded from the back and front of the head in different subjects. (A) E.D.A. Electrodes on vertex and occiput. (B) On vertex and forehead. (C) C.L.P. vertex and occiput. (D) Vertex and forehead. (E) W.H. vertex and occiput. (F) Vertex and forehead. In record B the small, rapid excursions are due to muscle action currents. Comparison of waves recorded from the back and front of the head in different subjects. (A) E.D.A. Electrodes on vertex and occiput. (B) On vertex and forehead. (C) C.L.P. vertex and occiput. (D) Vertex and forehead. (E) W.H. vertex and occiput. (F) Vertex and forehead. In record B the small, rapid excursions are due to muscle action currents. Diagrams showing current distribution. (A) If the skull was spherical. (B) In the median plane with the skull intact. (C) With an opening in the skull. Diagrams showing current distribution. (A) If the skull was spherical. (B) In the median plane with the skull intact. (C) With an opening in the skull. The relationship between the Berger rhythm and vision is clear. Nothing abolishes the activity more easily than a visual stimulus, however uninteresting that may be. More specifically, it is the recognition of pattern in the central field, or the attempt to perceive it, which interferes with the rhythm. Adrian and Matthews try various perturbations to show that the awareness of shape, not the level of illumination, most effectively stops the rhythm; pressure on the eyeball with the eyes closed may result in a circle of light that also abolishes the rhythm, and bright after-images delay its reappearance. Whereas sharp contrasts near the point of fixation cause the Berger rhythm to cease, those more peripherally placed are relatively, or absolutely, uninfluential. Activity appears within half a second after closing the eyes, but returns over several minutes if contrast is removed slowly by fading objects from vision with the eyes open. This difference is due to the abrupt withdrawal of attention with eye closure as compared to the continuing search for items of interest when the eyes are open but the visual world made blank. Also under these conditions, the rhythm persists more easily if the subject repeatedly opens the eyes expecting to see nothing based on immediate past experience, unless a deliberate effort is made to find something to see. In fact any intense effort directed at seeing something, even with the eyes closed, will quickly abolish the Berger rhythm. It is the change from one state to another that creates the most marked potentials and these reduce in amplitude if a steady state of eye closure is allowed to develop over a period of 10 min. In support of the critical contribution of pattern vision, Adrian and Matthews are unable to produce the Berger rhythm in any one of three individuals with long-standing blindness. Any form of intense concentration such as responding to questions, mental arithmetic, tying and untying knots, hearing, or registering unexpected touch abolishes the potentials, although subjects may habituate to these intrusions and return to the peace and quiet needed for their Berger rhythm to flourish. Conversely, muscular effort and less demanding intellectual activities—counting aloud, reciting poetry, chatting idly, or even Adrian listening to the amplified sound of his own Berger rhythm—have no effect on the potentials (Fig. 5). Abolition of the rhythm by non-visual activities. (A) E.D.A. Eyes closed. Problem in mental arithmetic given at signal. (B) W.H. Eyes closed. Touch on the nose with cotton wool. (C) Persistence of rhythm in spite of muscular effort. W.H. Eyes closed, squeezing pliers as tightly as possible. Abolition of the rhythm by non-visual activities. (A) E.D.A. Eyes closed. Problem in mental arithmetic given at signal. (B) W.H. Eyes closed. Touch on the nose with cotton wool. (C) Persistence of rhythm in spite of muscular effort. W.H. Eyes closed, squeezing pliers as tightly as possible. What causes the beat of these occipital neurons? The conditions of excitation must be uniform; and the synchrony results from their interconnections. Adrian and Matthews have already shown that rhythmic activity observed from the water beetle (Dytiscus marginalis) eye in darkness breaks up when retinal stimulation is non-uniform and of reduced brightness (and with all due modesty, the Nobel Laureate points out that his own Berger rhythm is more or less identical to that of the water beetle; Fig. 6 and see cover). In fact any new stimulation disturbs the cortical beat in vertebrates and invertebrates, replacing synchronous with asynchronous activity, as afferent signals arrive or widespread cortical activations invade the beating area. The further origin of the beat within the occipital cortex is less easy to place; it is not the striate area itself, and has nothing to do with light, but rather that part of the visual cortex directing attention to the amount of contrast provided by pattern recognition both for real and imagined objects. But Adrian and Matthews argue that this is not a rhythm that depends on the general (anatomical) properties of attention; instead it is something specific to visual attention. However, they acknowledge that nothing is likely to have more widespread an influence on the entire cortex than vision; they point out that other activities can intrude on the synchronous beat even in the absence of vision to the extent that the Berger rhythm disappears and it is not observed at all in the blind. Such differences as do exist between animals and man may depend on the superior visual apparatus of the latter; and they explain why non-visual activities more easily disrupt the Berger rhythm in cats and rabbits—notwithstanding electrical similarities between Professor Adrian and the water beetle—and the difference between recording directly from the cortical surface in animals and from the human scalp. This latter principle is demonstrated by studying the changes in Berger rhythm in a patient of Mr (Sir Hugh) Cairns studied peri-operatively before and within 90 min of removing a cerebral tumour. Again, while more extensive use of recordings from the cortical surface in man is desirable, the evidence that does exist is consistent with the view that the rhythm arises from the occipital cortex. Comparison of waves from water beetle preparation in darkness and light (upper record) and from human subject (E.D.A.) with eyes closed and open (lower record). In both, the rhythm is abolished during visual activity. Comparison of waves from water beetle preparation in darkness and light (upper record) and from human subject (E.D.A.) with eyes closed and open (lower record). In both, the rhythm is abolished during visual activity. Despite dependence of the Berger activity on the absence of attention to visual patterns, ‘flicker’ is able to induce rhythms that can occur at a frequency of up to 25 cycles/s. The apparatus designed by Adrian and Matthews provides a stimulus of medium light intensity that flickers across the entire visual field. Sitting in front of an opal glass bowl with his head covered by a black velvet curtain viewing a 30 watt bulb rotated by a gramophone motor, initially with his eyes closed and the stable 10 cycles/s Berger rhythm in place, Professor Adrian is exposed to the flickering light and ‘responds’ with a series of potential waves that are mostly synchronous with the flicker (Fig. 7). Ideally, these need to be in the range of 10–20/s: at slower rates, the rhythms are unstable; and if too fast, they may halve or double by comparison with frequency of the visual stimulus. This flicker response is more resilient to intrusions from other sensory stimuli or mental activity since ‘a coordinated beat is imposed on the area by the rhythmic excitation, whereas with the spontaneous waves there is nothing but their own interaction to synchronize the different neurones’. As to localization, the responsive area is thought to be sufficiently close to the striate region so as to be affected by its rhythmic pulsation, but not concerned directly with the perception of light. Examples of different rhythms induced by flicker. Subject E.D.A. signal line shows frequency of flicker. (A) Flicker at 8/s. (B) Eyes closed and field dark. Berger rhythm at 10/s. (C) Flicker at 12/s. (D) Flicker at 18/s. (E) Flicker at 10/s. Compare with B. Examples of different rhythms induced by flicker. Subject E.D.A. signal line shows frequency of flicker. (A) Flicker at 8/s. (B) Eyes closed and field dark. Berger rhythm at 10/s. (C) Flicker at 12/s. (D) Flicker at 18/s. (E) Flicker at 10/s. Compare with B. Although Adrian and Matthews have studied themselves (E.D.A. is an excellent subject whereas B.H.C.M. is slow to develop his Berger rhythm) and other laboratory workers (we have been unable to verify W.H., K.Y. and C.L.P.), Dr Berger has made recordings during a variety of altered physiological states and in the context of disease, detecting rhythms that are slower in these pathological conditions. In the main, the findings and conclusions of these investigators are in agreement; but they differ in one important respect. Adrian and Matthews find their observations on localization and flicker to be at odds with Dr Berger’s conclusion that the rhythm originates from every part of the cortex—afferent stimuli causing a more intense local pulsation that suppresses the beat over the rest of the brain so that it can no longer be detected outside the skull. Since vision dominates the work of the brain he argues that it is especially potent at disturbing the underlying rhythm. They do not accept that the rhythm is suppressed by attention to local afferent traffic but, alternatively, consider that it becomes asynchronous and so disappears. And the Berger rhythm is dissimilar to the beat of a pure receptor system such as the retina. However, they do agree that the interest of the Berger rhythm lies in its relation to attention, and coordinated activities of the cerebrum, albeit as a marker of inactivity emerging when the cortex ‘has nothing to do and disappearing as soon as the area resumes its normal work’. Taken with the many thousands of papers that have since used the electroencephalogram to study the human brain in health and disease, including several in the current issue, Adrian and Matthews’s conclusion that ‘the Berger rhythm shows the background of spontaneous discharge, but it can also show when that background is obscured and this may be information well worth having, from the practical as well as the theoretical aspect’ can be considered something of an understatement.
No takes yet. Share an insight, caveat, or question.
A. Compston (2010) studied this question.